Sorry you did not like the article Philipp but let me address your substantive comments. Kafka is most certainly a compute engine beyond a message broker. First, Kafka users routinely tune parallelism for consumer throughput. Additionally, users commonly deploy significant computation using the Streams API to allow an application to act as a stream processor, consuming an input stream from one or more topics and producing an output stream to one or more output topics, effectively transforming the input streams to output streams. Aggregations and window functions are common computations to derive rates and accelerations on changing sensor values for example.

With regard to combining workloads, we have to agree to disagree, but I suggest you take a peek at Gartner’s HTAP work and Forrester’s Translytical work suggesting there are many workloads that benefit from hybrid engines.